Analyse maisons - Shapash
Interpretation des predictions maisons
Project_Information
Author : VotreNom
Description : Rapport Shapash pour maisons
Project_Name : Analyse lightgbm_maison
Model analysis
Model used : LGBMRegressor
Library : lightgbm.sklearn
Library version : 4.6.0
Model parameters :
| Parameter key | Parameter value |
|---|---|
| boosting_type | gbdt |
| objective | None |
| num_leaves | 43 |
| max_depth | 9 |
| learning_rate | 0.1269191174033963 |
| n_estimators | 176 |
| subsample_for_bin | 200000 |
| min_split_gain | 0.0 |
| min_child_weight | 0.001 |
| min_child_samples | 10 |
| subsample | 1.0 |
| subsample_freq | 0 |
| colsample_bytree | 1.0 |
| reg_alpha | 0.0 |
| reg_lambda | 0.0 |
| random_state | None |
| Parameter key | Parameter value |
|---|---|
| n_jobs | None |
| importance_type | split |
| _Booster | |
| _evals_result | {} |
| _best_score | defaultdict( |
| _best_iteration | 0 |
| _other_params | {} |
| _objective | regression |
| class_weight | None |
| _class_weight | None |
| _class_map | None |
| _n_features | 55 |
| _n_features_in | 55 |
| _classes | None |
| _n_classes | -1 |
| fitted_ | True |
Dataset analysis
Global analysis
| Training dataset | Prediction dataset | |
|---|---|---|
| number of features | NaN | 55 |
| number of observations | NaN | 2,783 |
| missing values | NaN | 0 |
| % missing values | NaN | 0 |
Univariate analysis
- CODE_IRIS
- Revenu_fiscal_de_r__f__rence_par_habitant
- STATUT_COM_UU_Banlieue
- STATUT_COM_UU_Campagne
- STATUT_COM_UU_Ville isolée
- STATUT_COM_UU_Ville-centre
- TYP_IRIS_y
- UU2020
- anciennete_annonce
- annee
- annee_construction
- annonce_exclusive
- ascenseur
- bain
- balcon
- categorie_annonceur
- cave
- chauffage_energie
- chauffage_mode
- chauffage_systeme
- classe_annee
- dpeC
- dpeL
- eau
- etage
- exposition_autre
- exposition_est
- exposition_nord
- exposition_sud
- ges_class
- ges_class_encoded
- logement_neuf
- mapCoordonneesLatitude
- mapCoordonneesLongitude
- nb_etages
- nb_pieces
- nb_toilettes
- places_parking
- score__conomie_brut
- score__conomie_ratio_1000
- score_commerce_brut
- score_commerce_ratio_1000
- score_culture_loisir_brut
- score_culture_loisir_ratio_1000
- score_education_brut
- score_education_ratio_1000
- score_s_curit__brut
- score_s_curit__ratio_1000
- score_sant__brut
- score_sant__ratio_1000
- score_transport_brut
- score_transport_ratio_1000
- source_imputation
- surface
- surface_terrain
CODE_IRIS - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 222 |
| std | 133 |
| min | 0 |
| 25% | 102 |
| 50% | 222 |
| 75% | 339 |
| max | 448 |
Revenu_fiscal_de_r__f__rence_par_habitant - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 0.0159 |
| std | 1.02 |
| min | -1.72 |
| 25% | -0.704 |
| 50% | -0.177 |
| 75% | 0.426 |
| max | 3.66 |
STATUT_COM_UU_Banlieue - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
STATUT_COM_UU_Campagne - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
STATUT_COM_UU_Ville isolée - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
STATUT_COM_UU_Ville-centre - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
TYP_IRIS_y - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
UU2020 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0333 |
| std | 0.997 |
| min | -0.996 |
| 25% | -0.996 |
| 50% | -0.541 |
| 75% | 0.874 |
| max | 1.62 |
anciennete_annonce - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 5 |
| missing values | 0 |
annee - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 5 |
| missing values | 0 |
annee_construction - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 0.0223 |
| std | 0.98 |
| min | -4.93 |
| 25% | -0.375 |
| 50% | 0.122 |
| 75% | 0.677 |
| max | 1.58 |
annonce_exclusive - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
ascenseur - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 1 |
| missing values | 0 |
bain - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 8 |
| missing values | 0 |
balcon - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 4 |
| missing values | 0 |
categorie_annonceur - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
cave - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
chauffage_energie - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 5 |
| missing values | 0 |
chauffage_mode - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
chauffage_systeme - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 4 |
| missing values | 0 |
classe_annee - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 4 |
| missing values | 0 |
dpeC - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 0.00942 |
| std | 1.02 |
| min | -1.53 |
| 25% | -0.9 |
| 50% | -0.204 |
| 75% | 0.565 |
| max | 5.58 |
dpeL - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 7 |
| missing values | 0 |
eau - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 5 |
| missing values | 0 |
etage - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 1 |
| missing values | 0 |
exposition_autre - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
exposition_est - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
exposition_nord - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
exposition_sud - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
ges_class - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 8 |
| missing values | 0 |
ges_class_encoded - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 8 |
| missing values | 0 |
logement_neuf - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
mapCoordonneesLatitude - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0067 |
| std | 0.986 |
| min | -2 |
| 25% | -0.777 |
| 50% | -0.154 |
| 75% | 0.694 |
| max | 2.66 |
mapCoordonneesLongitude - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0175 |
| std | 0.995 |
| min | -2.94 |
| 25% | -0.632 |
| 50% | 0.0483 |
| 75% | 0.61 |
| max | 2.1 |
nb_etages - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 4 |
| missing values | 0 |
nb_pieces - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0127 |
| std | 0.986 |
| min | -2.38 |
| 25% | -0.357 |
| 50% | -0.357 |
| 75% | 0.149 |
| max | 7.24 |
nb_toilettes - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 8 |
| missing values | 0 |
places_parking - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
score__conomie_brut - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.00689 |
| std | 0.989 |
| min | -0.407 |
| 25% | -0.396 |
| 50% | -0.357 |
| 75% | -0.197 |
| max | 3.61 |
score__conomie_ratio_1000 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0115 |
| std | 0.978 |
| min | -1.27 |
| 25% | -0.791 |
| 50% | -0.131 |
| 75% | 0.55 |
| max | 6.25 |
score_commerce_brut - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0081 |
| std | 0.989 |
| min | -0.404 |
| 25% | -0.396 |
| 50% | -0.366 |
| 75% | -0.221 |
| max | 3.63 |
score_commerce_ratio_1000 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0296 |
| std | 0.941 |
| min | -1.33 |
| 25% | -0.826 |
| 50% | -0.188 |
| 75% | 0.58 |
| max | 5.48 |
score_culture_loisir_brut - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 0.00293 |
| std | 1.02 |
| min | -0.608 |
| 25% | -0.503 |
| 50% | -0.387 |
| 75% | -0.134 |
| max | 4.12 |
score_culture_loisir_ratio_1000 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.000241 |
| std | 1.05 |
| min | -0.656 |
| 25% | -0.542 |
| 50% | -0.341 |
| 75% | 0.0772 |
| max | 8.01 |
score_education_brut - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.00509 |
| std | 0.989 |
| min | -0.399 |
| 25% | -0.381 |
| 50% | -0.354 |
| 75% | -0.18 |
| max | 3.6 |
score_education_ratio_1000 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 0.0322 |
| std | 1.03 |
| min | -2.04 |
| 25% | -0.644 |
| 50% | -0.118 |
| 75% | 0.411 |
| max | 5.16 |
score_s_curit__brut - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 1 |
| missing values | 0 |
score_s_curit__ratio_1000 - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 1 |
| missing values | 0 |
score_sant__brut - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 8 |
| missing values | 0 |
score_sant__ratio_1000 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 0.00658 |
| std | 1.04 |
| min | -0.954 |
| 25% | -0.954 |
| 50% | 0.0667 |
| 75% | 0.388 |
| max | 4.2 |
score_transport_brut - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.00842 |
| std | 0.984 |
| min | -0.436 |
| 25% | -0.404 |
| 50% | -0.364 |
| 75% | -0.185 |
| max | 3.83 |
score_transport_ratio_1000 - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0106 |
| std | 1.01 |
| min | -6.72 |
| 25% | -0.729 |
| 50% | -0.145 |
| 75% | 0.605 |
| max | 3.38 |
source_imputation - Categorical
| Prediction dataset | |
|---|---|
| distinct values | 2 |
| missing values | 0 |
surface - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0184 |
| std | 1 |
| min | -1.64 |
| 25% | -0.619 |
| 50% | -0.31 |
| 75% | 0.229 |
| max | 7.86 |
surface_terrain - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | -0.0261 |
| std | 0.899 |
| min | -0.81 |
| 25% | -0.353 |
| 50% | -0.233 |
| 75% | 0.0151 |
| max | 9.52 |
Target analysis
prix_m2_vente - Numeric
| Prediction dataset | |
|---|---|
| count | 2,783 |
| mean | 2,710 |
| std | 951 |
| min | 150 |
| 25% | 2,100 |
| 50% | 2,700 |
| 75% | 3,250 |
| max | 8,190 |
Multivariate analysis
Model explainability
Note : the explainability graphs were generated using the test set only.
Global feature importance plot
Features contribution plots
- CODE_IRIS
- Revenu_fiscal_de_r__f__rence_par_habitant
- STATUT_COM_UU_Banlieue
- STATUT_COM_UU_Campagne
- STATUT_COM_UU_Ville isolée
- STATUT_COM_UU_Ville-centre
- TYP_IRIS_y
- UU2020
- anciennete_annonce
- annee
- annee_construction
- annonce_exclusive
- ascenseur
- bain
- balcon
- categorie_annonceur
- cave
- chauffage_energie
- chauffage_mode
- chauffage_systeme
- classe_annee
- dpeC
- dpeL
- eau
- etage
- exposition_autre
- exposition_est
- exposition_nord
- exposition_sud
- ges_class
- ges_class_encoded
- logement_neuf
- mapCoordonneesLatitude
- mapCoordonneesLongitude
- nb_etages
- nb_pieces
- nb_toilettes
- places_parking
- score__conomie_brut
- score__conomie_ratio_1000
- score_commerce_brut
- score_commerce_ratio_1000
- score_culture_loisir_brut
- score_culture_loisir_ratio_1000
- score_education_brut
- score_education_ratio_1000
- score_s_curit__brut
- score_s_curit__ratio_1000
- score_sant__brut
- score_sant__ratio_1000
- score_transport_brut
- score_transport_ratio_1000
- source_imputation
- surface
- surface_terrain
CODE_IRIS -
Revenu_fiscal_de_r__f__rence_par_habitant -
STATUT_COM_UU_Banlieue -
STATUT_COM_UU_Campagne -
STATUT_COM_UU_Ville isolée -
STATUT_COM_UU_Ville-centre -
TYP_IRIS_y -
UU2020 -
anciennete_annonce -
annee -
annee_construction -
annonce_exclusive -
ascenseur -
bain -
balcon -
categorie_annonceur -
cave -
chauffage_energie -
chauffage_mode -
chauffage_systeme -
classe_annee -
dpeC -
dpeL -
eau -
etage -
exposition_autre -
exposition_est -
exposition_nord -
exposition_sud -
ges_class -
ges_class_encoded -
logement_neuf -
mapCoordonneesLatitude -
mapCoordonneesLongitude -
nb_etages -
nb_pieces -
nb_toilettes -
places_parking -
score__conomie_brut -
score__conomie_ratio_1000 -
score_commerce_brut -
score_commerce_ratio_1000 -
score_culture_loisir_brut -
score_culture_loisir_ratio_1000 -
score_education_brut -
score_education_ratio_1000 -
score_s_curit__brut -
score_s_curit__ratio_1000 -
score_sant__brut -
score_sant__ratio_1000 -
score_transport_brut -
score_transport_ratio_1000 -
source_imputation -
surface -
surface_terrain -
Model performance
Univariate analysis of target variable
prix_m2_vente - Numeric
| True values | Prediction values | |
|---|---|---|
| count | 2,783 | 2,783 |
| mean | 2,710 | 2,700 |
| std | 951 | 762 |
| min | 150 | 355 |
| 25% | 2,100 | 2,220 |
| 50% | 2,700 | 2,710 |
| 75% | 3,250 | 3,180 |
| max | 8,190 | 6,340 |
Metrics
MAE : 402
R2 : 0.661
MSE : 306,000
MAPE : 0.172
MdAE : 305
Explained Variance : 0.661